. Author manuscript; available in PMC: 2017 Oct 1.

Published in final edited form as: Exp Clin Psychopharmacol. 2016 Aug 15;24(5):376–389. doi: 10.1037/pha0000088

Table 3.

Association of Conduct Problems to Substance Use-Related Outcomes and Mediation by Alternative and Complementary Reinforcement

	Total Effect	Component Paths		Mediation: CP → Mediator → Outcome

	CP → Outcome	CP → Mediator	Mediator → Outcome Controlling for CP	Indirect effect	Direct effect
	B (95% CI)	B (95% CI)	B (95% CI)	B (95% CI)	B (95% CI)
Outcome: Past Six-Month Use Status; Mediator: Alternative Reinforcers (Overall Sample; N = 3,202)
Model 1 (Unadjusted)	.96 (.81, 1.10)^†	−.18 (−.21, −.15)^†	−.13 (−.22, −.03)^**	.02 (.001, .05)^**	.95 (.81, 1.09)^†
Model 2 (Adjusted)^a	.96 (.82, 1.11)^†	−.17 (−.21, −.14)^†	−.14 (−.24, −.04)^**	.02 (.002, .05)^**	.95 (.82, 1.09)^†
Model 3 (Adjusted)^b	.69 (.58, .80)^†	−.14 (−.17, −.11)^†	−.11 (−.21, −.01)^*	.02 (.002, .03)^*	.68 (.57, .79)^†
Model 4 (Adjusted)^c	.68 (.56, .80)^†	−.13 (−.16, −.09)^†	−.12 (−.23, −.02)^*	.02 (.002, .03)^*	.67 (.55, .78)^†

Outcome: Frequency Use; Mediator: Complementary Reinforcers (Past Six-Month Substance Users; N = 340)
Model 1 (Unadjusted)	.44 (.34, .53)^†	.41 (.21, .60)^†	.18 (.08, .28)^***	.08 (.002, .17)^***	.36 (.23, .48)^†
Model 2 (Adjusted)^a	.44 (.35, .53)^†	.42 (.22, .61)^†	.20 (.09, .30)^***	.08 (.002, .17)^***	.36 (.25, .47)^†
Model 3 (Adjusted)^b	.34 (.24, .43)^†	.37 (.17, .56)^†	.18 (.08, .29)^***	.07 (.01, .15)^**	.27 (.14, .41)^†
Model 4 (Adjusted)^c	.34 (.26, .42)^†	.37 (.16, .59)^†	.19 (.14, .24)^†	.07 (.003, .15)^***	.27 (.19, .34)^†

Outcome: Frequency Use; Mediator: Alternative Reinforcers (Past Six-Month Substance Users; N = 833)
Model 1 (Unadjusted)	.52 (.46, .58)^†	−.28 (−.32, −.23)^†	−.46 (−.60, −.31)^†	.13 (.05, .22)^†	.39 (.33, .46)^†
Model 2 (Adjusted)^a	.53 (.46, .60)^†	−.25 (−.32, −.19)^†	−.44 (−.59, −.29)^†	.11 (.03, .22)^†	.41 (.33, .48)^†
Model 3 (Adjusted)^b	.38 (.31, .46)^†	−.19 (−.28, −.11)^†	−.38 (−.54, −.24)^†	.07 (.01, .18)^†	.30 (.23, .38)^†
Model 4 (Adjusted)^c	.41 (.35, .46)^†	−.18 (−.26, −.10)^†	−.35 (−.49, −.22)^†	.06 (.004, .16)^†	.33 (.26, .40)^†

Note. B (95%CI) = Parameter estimate for predictor from Generalized Estimating Equation with 95% confidence interval. CP = Conduct Problem.

Adjusted model is adjusted for demographic covariates, including ethnicity, sex, and parental education level.

Adjusted model is adjusted for peer substance use and positive urgency in addition to demographic covariates tested in ^aAdjusted.

Adjusted model is adjusted for depression, anxiety, and panic symptoms in addition to covariates tested in ^aAdjusted and ^bAdjusted.

p < .05,

^**

p < .01,

^***

p < .001,

^†

p < .0001.